I'm not feeding it so much as sharing context.
This was a first draft "memory dump" from Chatty (ChatGPT) as that model has the most comprehensive model of me.
I now have a set of contextual frameworks to feed to any AI to accelerate contextual alignment. Think of it as a very basic, inefficient Vulcan mind meld š
I do discuss social media posts with Chatty, but it generally doesn't store that information, it simply uses ongoing discussions to alter weight dsitritution and context.
As I move from public cloud AI to private LLM's I need to be able to transfer and delete my brain from any model I interact with, which helps accelerate alignment and functionality.
Here is a very brief snippet of my projects file for context:
## Writing / legacy
- Mikeās book: āWhy Bitcoin? A History and Guideā.
- Grandfatherās memoir: āHoly Stones and Salt Junkā (Royal Navy life); desired outcomes include structure improvements, chapter titles, foreword, photo section, KDP-ready 6x9 formatting.
- Constraint: Mike prefers analysis/editorial guidance, not direct rewrites of his own text.